AI tools your tech employees need to learn
We break down all the top AI tools that offer a value-add to your business: chatbots, coding assistants, image and documentation generators, and more.
Nov 22, 2023 • 9 Minute Read
- Public Sector
- IT Ops
- Engineering Leadership
- Team Development
- AI & Machine Learning
- Learning & Development
Seemingly overnight, the AI space has exploded, now brimming with tools for every possible task — be it code debugging, story crafting, or investment analysis. But it also presents a business headache: which of the countless tools are worth investing in for your staff training?
This article slices through all the buzz of what’s out there, and hones in on what your technologists can leverage to enhance productivity at your workplace. Whether you’re a software developer striving for code perfection, an IT personnel mapping out infrastructure, or a product manager eager to streamline endless meetings and note-taking, there’s something here for everyone.
Table of contents
The best AI Chatbots for your business
Let’s kick things off with the OG of AI that kicked off the current revolution: chatbots. Chatbots like ChatGPT, Bard and Bing are driven by large language models (LLMs), letting you have a very human-like conversation with a “friend” who knows way more than any of your actual friends. Like, it knows everything.
If you need any sort of text-based output—code, emails, language translation, explanations, technology comparisons, math and more—then an AI chatbot will give you what you need. Let’s look at the three market leaders.
The hands-down leader, ChatGPT, was created by OpenAI, and its latest, super-capable model is GPT-4 (though GPT-3.5 is also more than capable for most things you need). What makes it so amazing? Let’s ask.
Also, ChatGPT can now intake images, opening up a whole world of new possibilities, from identifying objects to solving math equations, writing code based on a rough website mockup to fixing a bike based on images of the bike and the tools you have. Audio is also supported through the mobile app, so you can now have a conversation with ChatGPT through voice instead of only typing.
Bard comes to us courtesy of Google, and was a response to OpenAI’s release of ChatGPT. This chatbot is based on the PaLM LLM, which replaced the original LaMDA family of LLMs.
In addition to the standard text-based conversation, Bard can also process and respond to images and audio easily through the UI.
Bing is Microsoft’s foray into the world of LLMs, leveraging OpenAI’s GPT models. It integrates real-time search (lest you forget that Bing was originally a search engine) with the superhuman capabilities of GPT models, giving you up-to-date results in a human-friendly format. It also cites its sources, and supports images through the UI. OpenAI leverages Bing for its own real-time searching, so there's some crossover here in functionality.
The best AI tools for writing, debugging and explaining code
Ever find yourself stuck, staring at incomplete code, scouring the cobwebs of your brain to remember what you need to do? Or looking at someone else’s code, trying to decipher what they’ve done? Or spending hours on Stack Overflow and Google trying to fix a bug?
Your life can be a lot easier with the help of an AI pair-programmer. Code completion tools like GitHub Copilot/Copilot X and others let you write, explain and debug code all from the comfort of your integrated development environment (IDE). (ChatGPT is equally capable in this area, though just in a separate interface.)
Let’s look at some of the options out there.
1. GitHub Copilot/Copilot X
Similar to GitHub Copilot, Tabnine offers code completion, but with more IDEs supported, and a Starter package that’s totally free (giving you short code completions of 2-3 words). Tabnine uses an in-house LLM to do its auto-completion, and your data and code are never shared, eliminating any privacy concerns.
3. Amazon CodeWhisperer
If you’re already doing a lot of work in the Amazon Web Services (AWS) ecosystem, then CodeWhisperer could be a no-brainer for a code completion tool. It supports a variety of languages and SDKs, but it also offers first-class support and code suggestions for the AWS APIs. You also get enterprise-ready benefits like robust security and administration that come with being part of the AWS family.
The best AI tools for streamlining code reviews and code analysis
If you work in a team environment, you likely spend a fair amount of time on code reviews and pull requests. You might also use some code scanning and analysis tools to write better code so that there are fewer things to deal with during code reviews.
Artificial intelligence tools can make these tasks easier too. It’s like having a second set of eyes scanning for potential issues, offering tips to fix them, and generally streaming your day-to-day workflow
1. Amazon CodeGuru Security
2. Synk Code
Snyk (powered by DeepCode AI, and pronounced “sneak”) allows you to scan, prioritize, and fix security vulnerabilities in your code, open source dependencies, container images, and Infrastructure as Code (IaC) configurations . It can search for things like memory leaks, dependency conflicts, runtime errors, incorrect API usage and more. It will then produce a report to help you analyze and prioritize what needs fixing.
3. What the Diff
After you’ve used some of the AI tools above to write and optimize your code, it’s time to check it in for review by a human. That’s where What The Diff can help speed things up. This tool can automatically write descriptions for pull requests, help refactor small issues found during the process, and can even send out notifications to stakeholders to keep them in the loop. It can also detect things like syntax errors, typos, unused variables and more.
The best AI image generation and diagramming tools
Gone are the days when art was confined to the canvas and brush; today's AI tools are redefining digital art. Whether you need artwork for a game you’re developing, a lifelike portrait, or just some abstract textures, these tools take pixels to the next level with just a little bit of textual input.
My mind was a little bit blown the first time I used Midjourney. It generates some really impressive results, all from a simple text prompt. “A sketch of a cat wearing sunglasses sitting at a laptop.” “A photorealistic man looking at his phone while walking in a big city.”
After producing four images, you have the option to upscale them (for better resolution) or create four more versions of a particular image. And I know you want to see that cat sitting at a laptop:
You access Midjourney through a Discord server, which can feel a little clunky at first, but once you get the hang of it, it’s really straightforward.
Another image generation tool, this one comes from our friends at OpenAI (DALL-E-3 is now available to ChatGPT Plus and Enterprise customers through the platform). The DALL-E interface is super easy to use (and can also integrate with Slack), and is a great way to explore this genre of generative AI tools. It seems to struggle a little with photorealistic images of humans (though so does Midjourney…just check out the human hands!), but hopefully that will improve with the latest version.
You might already know and love Miro for its diagramming, whiteboarding, collaboration and workflow capabilities. Well, they’ve added AI to the mix now too. Miro AI can do things like generate or summarize sticky notes, create mind maps, generate code from a widget, create user stories and acceptance criteria.
The best AI documentation tools
Everybody’s favorite part of software development: documentation! Am I right?
Even though the Agile Manifesto declares, “Working software over comprehensive documentation,” you still need some type of documentation. And what better task to hand off to AI.
Mintlify is becoming the fan-favorite in this space. In your IDE, highlight the code you want to document, Mintlify will analyze the code to understand it, then generate comprehensive documentation for your methods. It can then take that documented code and generate end-user documentation, driven by markdown files.
Next time you need to write an overview of your software, a how-to guide, help center pages or FAQs, head over to Scribe to get some help. Scribe works by capturing workflows and processes step-by-step, then creates a reference guide based on those steps. It’s easy to get started with a Chrome extension, and when you’re done, you can share the Scribe via a shareable link, a PDF, embedded HTML or other formats.
The best AI meeting assistant tools
Let’s face it: we all spend way too much time in meetings. You prep for the meeting, and may also take notes during the meeting. Then after the meeting, you have all the work of summarizing the meeting and documenting action items. AI tools are getting better and better at helping with some of this tedium.
Perhaps the most comprehensive tool in this space is Otter.ai. You can connect Otter to your Google or Microsoft calendar, and it then joins the meetings to record them in Zoom, Microsoft Teams or Google Meet. It will write notes, capture action items and generate summaries as well. And if you didn’t have Otter helping you in real-time, you can also give it an existing recording for a transcript and summary.
Mem helps you organize and make sense of your messy notes. Even if you’re the world’s best note-taker, over time, it probably becomes difficult to go back and find what you need, when you need it. Mem connects different parts of your notes by using AI to tag things. Search also becomes easier and more powerful too. You can just ask a question and have a conversation about your notes (much like you can in ChatGPT), you can use the search bar, or browse by tag.
That was a lot! The explosive growth and versatility of AI tools is remarkable. From advanced chatbots like ChatGPT, Bard, and Bing to the code completion tools such as GitHub Copilot and Tabnine, AI continues to revolutionize the way we work. We also saw how AI can streamline code reviews, create striking digital art, remove the tedium from documentation, and even make meetings a little less of a pain.
It’s an exciting space for sure, and it’ll be interesting to see what new-and-improved AI tools and capabilities we get in the future. If you want to dig deeper on some of the tools, check out these additional resources: